Selenium is an essential micronutrient in the diet of humans and other mammals. Many health benefits have been attributed to selenium that include preventing various forms of cancer (e.g., colon cancer, prostate cancer, lung cancer and liver cancer), heart disease and other cardiovascular and muscle disorders, inhibiting viral expression, delaying the progression of acquired immunodeficiency syndrome (AIDS) in human immunodeficiency virus (HIV)-positive patients, slowing the aging process, and having roles in mammalian development, including male reproduction and immune function. Numerous human clinical trails have been undertaken in recent years to assess the role of this element in cancer prevention, delaying the progression of AIDS, etc., at a cost of billions of dollars, but little is known about the mechanism of how selenium acts at the metabolic level in mammals to exert these many health benefits. We have proposed that the health benefits of selenium are due largely to its presence in selenoproteins as the selenium-containing amino acid, selenocysteine (Sec). Our program therefore focuses on: 1) developing mouse models to assess the role of selenium and selenoproteins in cancer prevention and development, 2) characterizing and elucidating the function of various selenoproteins and their roles in cancer prevention and development, and 3) identifying the means by which Sec is biosynthesized and incorporated into protein. The project discussed herein examines our research on the development of various mouse models for determining the role of selenium in cancer, cancer prevention and development. In the past several years, we have focused on completing our studies on characterizing mouse models: 1) that knockout the Sec tRNA gene (designated Trsp) and consequently result in the loss of selenoprotein expression in (a) T cells, (b) macrophage, (c) epidermal skin tissue, and (d) liver, and subsequent rescue or partial rescue of selenoprotein expression with a wild-type or mutant Trsp transgene. Our direction of focus in this project has shifted in the past year almost totally towards generating mouse models that deal with the role of selenoproteins in the progression of cancer and the targeted removal of individual selenoproteins instead of knocking out all selenoproteins through the removal of Trsp. In an earlier fiscal year, we completed our study on examining the molecular mechanism of selenoproteins in cell immunity and in the past year, we emphasized providing sufficient mice to complete a collaborative study with Dr. Melinda Beck in determining the effects of deficient, adequate, and supplemental levels of selenium intake on the immune response mediated by T cells in mice that do not express selenoproteins in T cells or express only the critical, housekeeping selenoproteins (but not the non-essential, stress-related selenoproteins). We have continued a study on targeting the removal of only thioredoxin reductase 1 (TR1) and/or glutathione peroxidase 4 (GPx4) in various tissues to examine the role of these two essential selenoproteins in several organs and tissues. Loss of GPx4 in epidermal tissue (designated skin herein) virtually mimicked the loss of all selenoproteins with the exception that the mice recover after day 10, when they routinely died in Trsp knockout mice, and appear to live a virtually normal life. The GPx4 knockout mice had alopecia along with a flaky and fragile skin and histological studies revealed epidermal hyperplasia along with changes in hair follicle appearance. These observations demonstrate a prior unknown role of GPx4 in cutaneous development providing further evidence that this selenoprotein has many essential functions in cell development. The targeted removal of TR1 in skin did not appear to have any alteration in phenotype. In our in vivo colon carcinogenesis study, we were able to demonstrate that Sep15 knockout mice, compared to wild type or heterozygote littermate controls, appeared to be protected against azoxymethane-induced formation of pre-neoplastic lesions (aberrant crypt foci). Furthermore, using microarrays as well as real-time rtPCR analysis of the colonic mucosa showed that guanylate-binding protein 1 was highly up-regulated in Sep15 knockout mice. The possible connection between this interferon-gamma-regulated protein and the 15kDa selenoprotein in colon cancer will be investigated in future experiments. In addition, the possible influence of dietary selenium in our colon cancer model using Sep15 knockout mice will be further elucidated. Furthermore, our other mouse study model, which involved using liver TR1 deficient mice exposed to the liver carcinogen, diethylnitrosamine, (DEN), came to fruition in the past year. We found that liver-specific TR1 knockout mice developed tumors (hepatic adenomas, hepatic carcinomas and hepatocholangiocellular adenomas) much more readily than control mice. 16 of 18 knockout mice developed tumors compared to 2 of 19 control mice. In addition, we found an upregulation of Nrf2-modulated genes in the TR1 deficient livers, including several of the glutathione transferases and the selenoprotein, glutathione peroxidase 2, which was even further upregulated in tumors.